1
Show HN: Rubiks Cube Solver
开源魔方全能平台，集成AI求解、计时与算法库，浏览器直接运行
Speedcube is an open-source platform for speedcubers featuring a Rubik's Cube solver, competition timer, algorithm library, and AI-assisted cube recog…
2
Structural Grid Descriptors Predict Within-Task Solver Success on ARC-AGI
用网格结构描述符预测符号求解器在ARC-AGI任务中的成功率，为抽象推理任务提供新视角
arXiv:2606.09026v1 Announce Type: new Abstract: We ask whether structural properties of intermediate grid states predict whether a symbolic ARC-AGI so…
3
Robust Asynchronous Planning via Auto-Formalization
探索LLM通过自动形式化将异步现实任务转化为可求解规划的新范式。
arXiv:2606.00981v1 Announce Type: new Abstract: LLMs can plan by either generating action sequences directly as a Planner or translating tasks into do…
4
PokerSkill: LLMs Can Play Expert-Level Poker without Training or Solvers
无需训练或求解器，大型语言模型即可在德州扑克中达到专家级水平，突破传统AI极限。
arXiv:2605.30094v1 Announce Type: new Abstract: Poker is a landmark challenge for artificial intelligence. The dominant approach relies on equilibrium…
5
An analytic theory of convolutional neural network inverse problems solvers
为CNN逆问题求解器提供严格的数学分析，揭示其工作原理与性能边界
arXiv:2601.10334v2 Announce Type: replace-cross Abstract: Supervised convolutional neural networks (CNNs) are widely used to solve imaging inverse pro…
6
阿里达摩院首次发布 GPU 版本求解器：稳定求解超 99% 问题类型，突破亿级变量“不可解”难题
IT之家 5 月 28 日消息，阿里巴巴达摩院今日宣布，“敏迭”求解器（MindOpt）正式发布 GPU 版本 ，充分利用 GPU 并行加速特性，引入新算法 突破“长尾效应”难题 。 针对约 2000 个通用算例的测试显示，敏迭可将 99% 以上的问题类型稳定求解至高精度，更能 支持传统上“不可解”…
7
阿里达摩院首次发布GPU版本求解器
36氪获悉，阿里巴巴达摩院“敏迭”求解器（MindOpt）正式发布GPU版本，充分利用GPU并行加速特性，引入新算法突破“长尾效应”难题。针对约2000个通用算例的测试显示，敏迭可将99%以上的问题类型稳定求解至高精度，更能支持传统上“不可解”的亿级变量线性规划问题。
8
Which Changes Matter? Towards Trustworthy Legal AI via Relevance-Sensitive Evaluation and Solver-Grounded Reasoning
法律AI可信度研究新突破，提出相关性敏感评估与求解器推理结合的评估框架。
arXiv:2605.26530v1 Announce Type: new Abstract: Legal reasoning requires distinguishing changes that matter from those that do not. Legal AI should re…
9
AutoSG: LLM-Driven Solver Generation Solely from Task Prompts for Expensive Optimization
仅凭任务提示，LLM自动生成专用求解器，让昂贵优化成本大降，效率飙升。
arXiv:2605.25658v1 Announce Type: cross Abstract: Expensive optimization tasks are ubiquitous in real-world applications, demanding highly specialized…
10
Memory-Guided Tree Search with Cross-Branch Knowledge Transfer for LLM Solver Synthesis
大模型结合记忆引导树搜索与跨分支知识迁移，自动合成组合优化求解器，突破传统方法瓶颈。
arXiv:2605.17539v2 Announce Type: replace Abstract: Combinatorial optimization (CO) underlies decision-making from logistics to chip design, where inf…